Skip to main content

Table 3 Crime-mobility models with ARIMA errors: estimates, specifications and GOF measures

From: Crime and COVID-19: effect of changes in routine activities in Mexico City

  Mobility coefficient (SE) Specification ME MAPE AICc Q (df) KPSS
All crimes 0.167 (0.014)*** ARIMA (1,1,1) (1,0,1) [7] 0.73 6.28 −2588.28 20.33 (15) 0.09
Violent robbery 0.304 (0.020)*** ARIMA (1,1,2) (1,0,1) [7] 0.18 8.81 −1774.99 16.81 (14) 0.13
Non-violent robbery 0.373 (0.020)*** ARIMA (0,1,2) (2,0,0) [7] 0.83 11.26 −1256.83 18.30 (15) 0.42
Robbery against residence -0.040 (0.037) ARIMA (0,1,1) (2,0,0) [7] 0.48 23.88 443.62 15.04 (16) 0.29
Serious violent crime (non-sexual) 0.080 (0.035)* ARIMA (1,1,1) (1,0,1) [7] 0.50 17.98 −210.57 21.96 (15) 0.15
Sexual violence 0.222 (0.047)*** ARIMA (1,0,1) w/linear trend 0.83 35.60 1342.43 21.79 (16) 0.17
Domestic violence 0.000 (0.027) ARIMA (1,1,1) (1,0,1) [7] w/linear trend 0.52 13.61 −824.66 22.19 (14) 0.10
VAW helpline calls 0.136 (0.033)*** ARIMA (1,1,1) (1,0,3) [7] 1.02 17.26 −245.83 22.52 (13)* 0.07
  1. Notes: Mobility coefficients in log-scale; \(^{***}p<0.001\), \(^{**}p<0.01\), \(^{*}p<0.05\)